Duet3D 1HCL - Closed Loop Controller Beta test

jay_s_uk

lirwin17

Sorry, I'm not 100% clear on the actual issue here. Thanks for providing your M122 output, from tuning error: 0x1f, it seems like no runtime tuning has happened. Similar to when an axis is un-homed, a driver that is untuned will not move, so if the issue is that the axis isn't moving then that would be the expected behaviour in this instance.
However, if you have run tuning, there is a deeper issue here.

Details of tuning can be found on this page. Specifically, the two 'what do I need to do?' sections (1, 2) are compulsory before the system will move (eventually these should be part of the online config tool, but not just yet for the beta test)

If you weren't aware of this page in the docs, that's really useful for us to know because it tells us we need to draw more attention to that page

Thanks!
Louis

supertb1

@lirwin17 When I manually try to tune the X axis which is on closed loop mode on my config with everything selected it gives this error:
M569.5 P50.0 S500 A0 R0 D8191 V71
Error: M569.5: Drive is not in closed loop mode
M569.5 P50.0 S500 A0 R0 D8191 V71

Here is my config:
G4 S3 ;wait for expansion boards to start
M569.1 P50.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 50 with a quadrature encoder on the motor shaft that has 20 steps per motor full step.
M569.1 P51.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 51 with a quadrature encoder on the motor shaft that has 20 steps per motor full step.
M569 P50.0 D4 S1 ; Configure the motor on the 1HCL at can address 50 as being in closed-loop drive mode (D4), Open loop (D2) and not reversed (S1) X axis
M569 P51.0 D2 S1 ; Configure the motor on the 1HCL at can address 51 as being in closed-loop drive mode (D4), Open loop (D2) and not reversed (S1) Y axis
M569 P0.0 S1 ; physical drive 0.0 goes forwards Z 1-2 axis
M569 P0.4 S1 ; physical drive 0.4 goes forwards Extruder
M584 X50.0 Y51.0 Z0.0:0.1 E0.4 ; set drive mapping
M671 X-122.25:689.20 Y0:0 S3.0 ; leadscrews at left (connected to Z0) and right (connected to Z1) of X axis
M350 X32 Y32 Z32 E16 I1 ; configure microstepping with interpolation
M92 X106.5 Y257 Z795.00 E339.00 ; set steps per mm
M566 X900.00 Y900.00 Z60.00 E120.00 ; set maximum instantaneous speed changes (mm/min)
M203 X6000.00 Y6000.00 Z180.00 E1200.00 ; set maximum speeds (mm/min)
M201 X500.00 Y500.00 Z20.00 E250.00 ; set accelerations (mm/s^2)
M906 X1000 Y3000 Z3000 E800 I30 ; set motor currents (mA) and motor idle factor in per cent
M84 S30 ; Set idle timeout
M917 X0 Y0 ; Set the closed loop axes to have a holding current of zero

supertb1

@lirwin17 and it fails to manually tune

supertb1

@lirwin17 with just raw encoder reading and step maneuver it throws fault: M569.5 P50.0 S500 A0 R0 D1 V64 and maintains that the motor is not in closed loop mode

supertb1

@lirwin17 with just current motor steps and target motor steps selected performing the step maneuver yields: M569.5 P50.0 S500 A0 R0 D6 V64 and maintains that the drive is not in closed loop mode

T3P3Tony

@supertb1 i think you first need to tune the motors

M569.6

as outlined here:
https://duet3d.dozuki.com/Wiki/Tuning_the_Duet_3_Expansion_1HCL#Section_Tuning

If you are using a quadrature encoder (i.e. your M569.1 command includes T1 or T2), then from the table above, manoeuvres 1, 2, 4, 8 and 16 are required - this means that they must be run every time the printer is powered on. This can be achieved by using the following command:
M569.6 P##.# V31 ; Where P##.# is the driver address to tune

lirwin17

@supertb1
Hmm there are are a few M122 outputs so I've lost track a little...

The first ones show Closed loop enabled: yes, but then some later ones show no.

The problem seems to be that when you come to run the tuning move, that yes has flipped to a no...

Please could you run through the following:

power up (so config.g runs)
run a manoeuvre command - it might be best to start with the simplest one V1

And at each point check M122 to see when yes flipps to no. Is it showing yes after config.g has run, or is it only when the move is run that it flips to no?

supertb1

@lirwin17 That yes to the closed loop enable was there before I updated the 1HCL board to the latest firmware and now at power up it reports "no" to the close loop enable

supertb1

@lirwin17 9/20/2021, 4:20:18 PM M122 B50
Diagnostics for board 50:
Duet EXP1HCL firmware version 3.4.0beta3+1 (2021-09-14 15:07:31)
Bootloader ID: SAME5x bootloader version 2.4beta (2021-07-23)
Never used RAM 51580, free system stack 2594 words
Tasks: Move(notifyWait,0.0%,160) HEAT(notifyWait,0.0%,108) CanAsync(notifyWait,0.0%,70) CanRecv(notifyWait,0.0%,81) CanClock(notifyWait,0.0%,72) TMC(notifyWait,30.6%,358) CLSend(notifyWait,0.0%,148) CLData(notifyWait,0.0%,142) MAIN(running,67.4%,414) IDLE(ready,0.0%,39) AIN(notifyWait,2.0%,265), total 100.0%
Last reset 00:02:47 ago, cause: power up
Last software reset time unknown, reason: HardFault bfarValid precise, available RAM 51580, slot 0
Software reset code 0x0060 HFSR 0x40000000 CFSR 0x00008200 ICSR 0x00430803 BFAR 0x20030008 SP 0x2001eba8 Task MAIN Freestk 531 ok
Stack: 00000000 00000000 00000002 20030000 00030f39 0001b8f9 0001ea4c 41000000 bf800000 00000000 533c0605 04001200 a808a700 06420000 c3210828 19a21429 50708239 5d000040 2a60041c 5dd40122 4143051f 41200000 41c22c02 4142f6d6 20000010 0002699b 00000002
Closed loop enabled: no, live status: 0, encoder type none, pre-error threshold: 0.000000, error threshold: 0.000000, reverse polarity: no, tuning: 0, tuning error: 0, collecting data: no, ultimateGain=0.000000, oscillationPeriod=0.000000, Control loop runtime (ms): min=0.000000, max=0.058667, avg=0.000000, Control loop frequency (Hz): min=7281.553223, max=18292.683594, avg=17441.861328
Driver 0: pos 0, 106.5 steps/mm, standstill, reads 42667, writes 19 timeouts 0, SG min/max 0/0, steps req 0 dMoves scheduled 0, completed 0, in progress 0, hiccups 0, step errors 0, maxPrep 0, maxOverdue 0, maxInc 0, mcErrs 0, gcmErrs 0
Peak sync jitter -5/4, peak Rx sync delay 181, resyncs 0/0, no step interrupt scheduled
VIN: 24.3V, V12: 12.2V
MCU temperature: min 27.4C, current 29.5C, max 29.5C
Ticks since heat task active 230, ADC conversions started 164966, completed 164965, timed out 0, errs 0
Last sensors broadcast 0x00000000 found 0 235 ticks ago, loop time 0
CAN messages queued 1384, send timeouts 0, received 1509, lost 0, free buffers 37, min 37, error reg 0
dup 0, oos 0/0/0/0, bm 0, wbm 0, rxMotionDelay 0

supertb1

@lirwin17 I just reinstalled the latest firmware and it is still reporting no to close loop enable at power up

supertb1

@lirwin17 I tried rolling back the 1HCL board firmware to the previous one, still showing no on the enabled... and it won't let me do any of the tuning variables

supertb1

@t3p3tony I tried to change the closed loop back to open loop on the card in question and taking the "C" value out of " M569.1 P50.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 50 with a quadrature encoder on the motor shaft that has 20 steps per motor full step.
M569.1 P51.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 51 with a quadrature encoder on the motor shaft that has 20 steps per motor full step. and now with the encoded motor it will not operate in open mode either... I reinstalled the 3+1 firmware and put everything back the way it was....

supertb1

@lirwin17 Just for my information, when you say Power up (so config.g runs) you mean simply energizing the boards correct... or is there another command I need to enter to get the "config.g" to run?

supertb1

@supertb1 so after I did this using just the config it dawned on me that I probably would have to disconnect the encoder from the 1HCL board as well... you guys did a great job on these boards... very smart...lol

supertb1

@lirwin17 I'm not trying to be extra special here... I guess I need to start from the start... how do I trouble shoot this issue I am having?

T3P3Tony

@supertb1

I think you are using M569.5 instead of M569.6

What does sending

M569.6 P50.0 V31

Result in?

lirwin17

@supertb1

it dawned on me that I probably would have to disconnect the encoder from the 1HCL board as well... you guys did a great job on these boards... very smart...lol

Hmm you should never need to disconnect the encoder from the board unless you want to - the board should work in open-loop mode without anything needing to be disconnected, and you shouldn't need to disconnect it to cycle power or anything.

That yes to the closed loop enable was there before I updated the 1HCL board to the latest firmware and now at power up it reports "no" to the close loop enable

Righty I'm just looking over your config.g, it's been posted a few times, so let me just re-post here so that we're all looking at the same thing:

G4 S3 ;wait for expansion boards to start
M569.1 P50.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 50 with a quadrature encoder on the motor shaft that has 20 steps per motor full step.
M569.1 P51.0 S1 T2 C20 E20 R100 I0 D0 ; Configure the 1HCL board at CAN address 51 with a quadrature encoder on the motor shaft that has 20 steps per motor full step.
M569 P50.0 D4 S1 ; Configure the motor on the 1HCL at can address 50 as being in closed-loop drive mode (D4), Open loop (D2) and not reversed (S1) X axis
M569 P51.0 D2 S1 ; Configure the motor on the 1HCL at can address 51 as being in closed-loop drive mode (D4), Open loop (D2) and not reversed (S1) Y axis
M569 P0.0 S1 ; physical drive 0.0 goes forwards Z 1-2 axis
M569 P0.4 S1 ; physical drive 0.4 goes forwards Extruder
M584 X50.0 Y51.0 Z0.0:0.1 E0.4 ; set drive mapping
M671 X-122.25:689.20 Y0:0 S3.0 ; leadscrews at left (connected to Z0) and right (connected to Z1) of X axis
M350 X32 Y32 Z32 E16 I1 ; configure microstepping with interpolation
M92 X106.5 Y257 Z795.00 E339.00 ; set steps per mm
M566 X900.00 Y900.00 Z60.00 E120.00 ; set maximum instantaneous speed changes (mm/min)
M203 X6000.00 Y6000.00 Z180.00 E1200.00 ; set maximum speeds (mm/min)
M201 X500.00 Y500.00 Z20.00 E250.00 ; set accelerations (mm/s^2)
M906 X1000 Y3000 Z3000 E800 I30 ; set motor currents (mA) and motor idle factor in per cent
M84 S30 ; Set idle timeout
M917 X0 Y0 ; Set the closed loop axes to have a holding current of zero

The first thing that jumps out is line 5:

M569 P51.0 D2 S1 ; Configure the motor on the 1HCL at can address 51 as being in closed-loop drive mode (D4), Open loop (D2) and not reversed (S1) Y axis

Your comment says it's closed loop, but the actual command is using D2, which is an open loop mode.
However, your line above using P50.0 is correct, and that seems to be the driver you have been using, so I assume you are aware of this and this is all fine.

It might be worth only having one closed loop board plugged in at a time whilst we're diagnosing this issue. Just to rule out issues like them accidentally sharing a CAN address or something like that.

I suspect the problem here is line 2:

M569.1 P50.0 S1 T2 C20 E20 R100 I0 D0

This might be our fault - that particular command has gone through a number of changes, and it should have been updated in all the docs - except we seem to have missed this section! So if you've copy-pasted from that, this may be where our issues are coming from - I'll update that page now.
The GCODE dictionary page for M569.1 is definitely correct though, so I'll re-write your M569.1 command from first principles:

Your motor driver is 50.0 - so P50.0
Your encoder is a quadrature on motor shaft - so T2
From googling your motor serial number, you have this motor? That page gives a 'resolution' of 1000. No units though, how helpful. I'll assume that's CPR. That motor is also 200 step/rev, so we do 1000 ÷ 200 = 5 - so C5
Sensible error thresholds are maybe 5 & 10 - so E5:10
We'll leave R,I,D as default so you don't need to include them for now

Putting that all together:

M569.1 P50.0 T2 C5 E5:10

I suspect when your config.g file was running, it was erroring at your M569.1 command - so never getting to M569 to turn the 'no' to 'yes' in M122.

It might be worth running M98 P"config.g" manually and seeing if any errors come back. Then, change your M569.1 command for my M569.1 command, run M98 P"config.g" and hopefully no errors will come back.
Once that's done, turn it off and on again, and you should be good to go

lirwin17

@supertb1 @t3p3tony

Edit: I got my .5 and .6s confused I've corrected them now

Yes, Tony is also right here:

I think you are using M569.5 instead of M569.6
What does sending
M569.6 P50.0 V31
Result in?

The M569.5 command is for recording. This does let you pass in a manoeuvre (V parameter) to do. So running M569.5 as you describe would technically work to do a tuning move...
However, you'll also be making a recording each time, so you'll quickly fill up your SD card with recordings. For this reason, unless you intend to do a recording, using M569.6 is preferable to do a tuning move.

For anyone curious as to why there are 2 GCODE commands to do tuning moves, the rationale is that with HTTP delays etc, it is unlikely that you'd be able to send M569.5 and M569.6 quickly enough to record a tuning move. For that reason, M569.5 allows you to specify a tuning move that gets started just before the recording is started.

supertb1

@t3p3tony said in Duet3D 1HCL - Closed Loop Controller Beta test:

M569.6 P50.0 V31

9/21/2021, 7:00:28 AM M569.6 P50.0 V31
Error: M569.6: Drive is not in closed loop mode.